Genome-wide Investigation of Alcohol Response: A Meta-Analytic Review and Polygenic Associations with AUD

(7)项目摘要/摘要

(7) Project Summary/Abstract Specific Aims The current proposal first aims to extend previous genetic studies of level of response (LR) to alcohol by conducting the largest genome-wide association study (GWAS) of LR to date through the meta-analysis of multiple samples with extant SRE (Self-Rating of the Effects of Alcohol) and GWAS data. A second aim is to use summary data from the described GWAS of LR to create polygenic risk scores in an independent sample in order to determine whether, and to what extent, the genetic influences underlying LR to alcohol serve as a risk factor for AUD. Method Towards the abovementioned aims, datasets will be processed according to standard quality control (QC) procedures (e.g., Anderson et al., 2010) allowing for genotype imputation to a common reference panel and GWA analysis of each dataset using methods appropriate for the individual study designs. For all identified samples (see attached contribution letters) Dr. Marc Schuckit (consultant) will provide critical support in decisions regarding the use of the SRE phenotype. All data-analytic procedures will be conducted in coordination with Dr. Ian Gizer (sponsor), Dr. Arpana Agrawal (consultant), and the respective dataset contributors to ensure that all data are analyzed according to mutually-agreed upon procedures. Following individual study-level GWA analyses, standardized GWAS meta-analysis QC procedures will be carried out (Winkler et al., 2014) and a meta-analysis combining results from all samples will be conducted utilizing a fixed-effects model in METAL (Willer et al., 2010). Lastly, results from the SRE meta-analysis will be utilized to compute polygenic risk scores (PRS) in an independent target sample to examine the predictive ability of the LR to alcohol PRS for AUD outcomes. The Australian Twin Families (OZ-ALC GWAS) sample accessed from the database of Genotypes and Phenotypes (dbGaP) will be used to accomplish this second aim. Long-Term Objectives The over-arching goal of the current proposal is to utilize GWA meta-analytic procedures and polygenic modeling approaches to synthesize quantitative data across multiple GWA datasets in order to advance our understanding of the genetic architecture of AUD etiology, as well as a well-established biological risk mechanism of AUD development (i.e., LR to alcohol). Through the combination of multiple genetically-informed datasets, the current study is well-positioned to examine the genetic factors underlying a well-defined endophenotype of AUD, LR to alcohol. Additionally, polygenic prediction models will serve to expand our knowledge of the polygenic architecture underlying LR to alcohol, as well as the shared genetic contributions between LR to alcohol and AUD. By gaining a more precise understanding of the genetic architecture of LR, and the subsequent susceptibility for AUD, we can make substantial progress towards integrating this genetic information in treatment approaches with the goal of developing personalized AUD intervention efforts. Training Aims The current proposal will allow the applicant to gain valuable experience in management, integration, and data analysis procedures for investigating alcohol use outcomes in large-scale, genetically-informed datasets, as well as knowledge in the application of state-of-the-art polygenic modeling approaches for examining the aggregate effect of genetic variation across the genome. The nature of these training experiences will provide an immersive experience for the applicant through close collaboration with multiple research groups. Thus, successful completion of these training aims will provide the foundation for the applicant’s career objective of obtaining a tenure-track academic position investigating AUD phenotypes in large, consortia-derived, genetic datasets and how best to leverage findings from such studies to improve clinical outcomes.